Speedbumps, Potholes, and Detours on the Road to Social Security Reform

As Yogi Berra once said, “You’ve got to be careful if you don’t know where you’re going ’cause you might not get there.” This year’s round of fuzzy congressional initiatives to “save” Social Security appears more likely to lose a golden opportunity to permanently improve the future retirement security of working Americans – unless current rhetoric and strategies are overhauled.

Last month, several days of House Ways and Means committee hearings reinforced how much the retirement program reform debate remains stalled and misdirected. With a few exceptions, the Social Security proposals unveiled on Capitol Hill share two common flaws. First, they treat the issue primarily as an effort to avoid future federal budget deficits and to preserve a positive balance of paper IOUs in the retirement program’s Trust Fund account. Second, the proposed bills strive too much to “re-guarantee” the inadequate retirement benefits promised under current law, instead of allowing greater opportunity for workers to seek higher returns by accepting the market risks of more volatile private investments.

The net effect is to blur the fundamental choice facing both policymakers and their constituents. Why should they undertake the tasks of sorting out new tradeoffs and engineering complex structural adjustments in retirement program mechanics if the overall objective fails to reach far beyond restoring the status quo promised under current law?


Archer-Shaw Plan

Consider the leading example of this cramped vision of Social Security reform — the “Social Security Guarantee Plan” advanced by House Ways & Means chairman Rep. Bill Archer (R-TX) and Social Security subcommittee chairman Clay Shaw (R-FL). Upon introducing the plan in late April, Archer proclaimed that its first and foremost principle was “personal security,” and he then explained “that is why this plan fully preserves the current Social Security system … for all generations to come.” (italics added for emphasis). The plan’s commitment to preserve the Social Security Trust Fund and the guarantees promised under the current retirement necessarily leads to the plan’s other fundamental feature. “There is no privatization of the system at all,” concluded an April 28 Ways & Means committee summary of the plan.

The Archer-Shaw plan essentially utilizes the device of “temporary” individual investment accounts to funnel additional general revenues into the current retirement program and prop up its underfinanced set of fixed benefit promises. The federal government would begin by reducing projected federal budget surpluses and using those funds to create an annual, refundable tax credit for all workers who pay payroll taxes into the Social Security retirement program. The tax credits would equal two percent of taxable earnings up to the Social Security wage cap ($72,600 in 1999), which would amount to a maximum credit of $1,452 under Archer-Shaw this year.

But few, if any, workers could expect to receive increased retirement benefits and higher overall rates of return on their payroll taxes from the deposits made to their Social Security Guarantee (SSG) accounts. Upon retirement, an individual worker’s SSG account would be reclaimed (read “confiscated”) by the Social Security Administration and converted to a monthly annuity for life. Under the Archer-Shaw plan’s 100-percent “clawback” provision, the SSG accounts first must replace, dollar-for-dollar, the monthly benefit checks that the retiree is promised under today’s Social Security program. Only workers whose SSG account balance grew large enough to finance a monthly annuity benefit greater than current law benefit promises would come out ahead.

That result remains highly unlikely under the Archer-Shaw plan’s design features. The SSG plan limits individual account deposits to less than 19 percent of current Old Age and Survivors Insurance (OASI) payroll taxes, and it requires that 40 percent of individual account funds be invested in relatively low-yielding corporate bonds. “One Man’s Ceiling Is Another Man’s Floor,” according to a 1973 Paul Simon song, but the Archer-Shaw plan ensures that current law retirement benefits will be the former, not the latter, for future workers. Even if individual account investment options were liberalized in future years, the SSG plan’s other features (guaranteed payment of promised OASI benefits, low limits on contributions to individual accounts, and 100-percent clawback) would create dangerous moral hazard incentives for workers to make one-way bets on high-risk investments.

Even if workers could partly escape the unyielding grip of the clawback provision, their Archer-Shaw accounts still would lack another key element of private ownership. Workers could not count on being able to pass on the wealth accumulated in their SSG account balances to their families and other heirs – unless they were “lucky” enough to die before retiring. Under the Archer-Shaw rules for retired workers, any money remaining in their individual account balances at death would be taken by the Social Security Administration — first to finance payment of their survivors’ benefits and then to “replenish” the OASI trust fund. Thus, retired workers who did not live as long as average retirees would lose twice – they would receive smaller total lifetime annuity payments, and their estates could not benefit from the nest egg of wealth they had accumulated while working.

In reality, workers would merely be “renting” the funds in their SSG individual accounts, not owning them. Because the Archer-Shaw plan’s policy imperative is to preserve at all costs the OASI Trust Fund ledger and the guaranteed benefits promised by current Social Security law, it sacrifices bequest incentives, snaps a short leash on potential private investment returns, and loses sight of a more important objective — improving retirement income prospects for working Americans.

By using additional general tax revenues to deliver essentially the same level of benefits promised by current law, the Archer-Shaw plan imposes the equivalent of an income tax increase that amounts to two percent of taxable payroll. After all, budget “surpluses” still represent taxpayers’ money. Propping up the current Social Security system with such funds incurs opportunity costs – in the form of better policy alternatives foregone, such as tax cuts or transitions to fundamental retirement reform. Moreover, in the not unlikely event that projected long-term budget surpluses actually fail to materialize, the Archer-Shaw commitment to individual account tax credits would require new and more transparent revenue sources (tax increases). Thus, today’s “bad deal” offered by Social Security would not get any better under Archer-Shaw, and workers’ overall rates of return on their higher tax payments for retirement benefits could easily worsen.

The Archer-Shaw plan differs from President Clinton’s Social Security financing proposal more in form than substance. Both proposals would use additional general tax revenues to cover future shortfalls in OASI funds needed to pay benefits. The Clinton plan has been lambasted for advocating a greater role for government-directed investments in private sector equities, by using a portion of “surplus” assets in the OASI trust fund. But Archer-Shaw blurs this point of criticism, because its own investment rules offer mostly distinctions without a difference. The ultimate “beneficiary” under both investment schemes is the current defined benefit structure for Social Security and the accompanying OASI Trust Fund balance. (Or, as Baby Boomer rock fans of The Who’s classic lyrics might recall, “Meet the new boss. Same as the old boss.”) Although Archer-Shaw would incur the expense of structuring individual accounts and encouraging service competition among as many as 50 private fund managers, the plan would prevent any significant variation in investment strategies and, in the final analysis, return virtually all investment gains back to the current law Social Security program. Why bother? Its so-called private investment accounts would be managed under government rules, with OASI’s budget account balance as the primary beneficiary. Even the Clinton plan offers more of an upside gain potential for low- to moderate-income savers, in the form of voluntary USA retirement accounts that would allow workers to capture private investment returns without an OASI clawback.

Kolbe-Stenholm Plan

Perhaps the most carefully developed bipartisan Social Security reform plan presented before the House Ways and Means Committee last month was the 21st Century Retirement Act sponsored by Rep. Jim Kolbe (R-AZ) and Rep. Charles Stenholm (D-TX). The proposed legislation (H.R. 1793) would require a “carve out” of two percent of taxable payroll for all workers below the age of 55 by the end of this year. Those workers’ payroll taxes would be diverted to create personal Individual Security Accounts (ISAs). But the focus of Kolbe-Stenholm is on restoring long-term fiscal solvency to the OASI Trust Fund and the unified federal budget, rather than substantially enhancing the retirement income prospects of working Americans.

Kolbe and Stenholm told the Ways & Means Committee on June 9 that their plan uses advance funding of ISAs to reduce the future unfunded liabilities of the current law Social Security program. They emphasized that those OASI Trust Fund liabilities, which begin to mount in 2014, would represent a claim on general revenues and future taxpayers, as well as ever-growing budgetary pressure that would crowd out other competing priorities. The carve out provisions in H.R. 1793 are aimed at shrinking future Trust Fund balances. Kolbe and Stenholm pointed out that “add on” individual account proposals, on the other hand, would take funds from today’s general revenues rather than the current payroll tax, simply allow the OASI Trust Fund to accumulate more IOUs as “assets,” and then leave too great a financial burden on future taxpayers to provide future retirement benefits.

In order to finance its two-percent carve out while reducing more than half of the future general fund liabilities between 2014 and 2034, Kolbe-Stenholm proposes a number of reductions in future retirement benefits growth. It would change the benefits formula to target middle- and upper-income workers (by adjusting bend points to lower levels), accelerate increases in the normal retirement age for full benefits, impose greater actuarial reductions for early retirement benefits, gradually phase in a modest increase in the eligibility age for early retirement, and increase the computation period for determining a worker’s earnings credited to Social Security.

Although H.R. 1783 imposes some “hard choices” in terms of future reductions in promised benefits and refrains from double counting contingent budget resources, Kolbe and Stenholm stretch the semantics of federal fiscal policy in claiming that their plan does not draw upon future budget surpluses. Their Ways & Means testimony admits elsewhere that “our plan uses the current Social Security surplus to create individual accounts that will provide benefits for future retirees” (italics added for emphasis). That “surplus” represents the projected amounts by which annual OASI payroll tax revenues will continue to exceed benefits outlays until 2014.

In addition, Kolbe-Stenholm adopts several questionable policy measures aimed at generating additional general budget resources for Social Security reform. It imposes a somewhat arbitrary “temporary” reduction of 0.33 percent in annual consumer price index (CPI) calculations, in order to trim Social Security benefit payments. More troubling, the CPI adjustment is used to “recapture” for the OASI Trust Fund both the revenue gains and spending reductions it provides in other parts of the federal budget. That provision sanctions an abandonment of full inflation indexing of the federal income tax and a return to the danger of bracket creep that moves taxpayers’ marginal tax rates higher over time.

H.R. 1783 also proposes to reclaim for OASI the portion of revenues from income taxes on Social Security benefits that currently is allocated to the Medicare Hospital Insurance Trust Fund. This revenue shift that rearranges the deck chairs on the entitlement Titanic will only accelerate Medicare’s own fiscal problems. It will not change the unified federal budget balance.

Kolbe-Stenholm tries to soften the sting of its proposed retirement benefit cuts and to increase the political appeal of its legislative package by providing several “progressive” benefit guarantees and enhancements. However, the combination of minimum benefit guarantees and ISA contribution assistance that H.R. 1783 provides to lower income workers sends a confusing message. It reinforces political commitments to defined benefits and income redistribution, and it raises doubts about embracing a more aggressive transition to defined contribution-based personal investment accounts for all workers.

H.R. 1783 further undercuts the rationale for its proposed retirement benefits reductions by failing to link them more closely to the structure and amount of an individual worker’s carve-out contributions to ISAs. Kolbe-Stenholm does not come close to matching the level of a worker’s payroll taxes diverted into an ISA with a proportionate offset in future benefits promised under current law Social Security. Instead, the bill’s benefit cut provisions look like a political grab bag of whatever nips and tucks can be surgically made in different components of the existing benefit formula, without attracting too much concerted attention and heated opposition. The bill’s sponsors may promise that most workers will come out ahead as the timing of benefit cuts is matched by growth in individual account balances. But Kolbe-Stenholm’s budget-driven approach to benefit cuts will leave many workers confused and uncertain, as they guess whether they will be winners or losers under such a complex menu of changes.

Thus, Kolbe-Stenholm makes the case for advancing to full privatization more difficult. Its financing structure essentially allows it only to afford as high a level of ISA contributions as it can squeeze out of OASI defined benefit promises. It fails to consider another key source of net national saving — reductions in growth of other federal government spending programs – which could be used to increase the funding base available for ISAs. Moreover, Kolbe-Stenholm’s restrictive plan for a handful of centralized investment options for ISAs diminishes their “ownership” appeal. The bill’s lead sponsors even touted the “virtue of hybridization” before the House Ways & Means Committee, pointing out that existing long-term Social Security liabilities and the needs of special populations prevent any move to full privatization.

The overall message sent by the bill’s cautious first step toward personal retirement accounts is that H.R. 1783 is much more concerned with balancing the books of the federal budget and the OASI Trust Fund than with harnessing the full potential of private markets and personal responsibility to secure a dramatically more prosperous retirement future for working Americans.

Bipartisan Senate Plan

The Bipartisan Social Security Reform Plan (sponsored by Sen. Judd Gregg (R-NH), John Breaux (D-LA), and five other senators) represents a less detailed Senate counterpart to Kolbe-Stenholm, with a few variations. It too provides a carve out from current Social Security payroll taxes, by diverting amounts equal to two percent of OASI-covered wages into new personal retirement savings accounts. However, it would require all workers, regardless of their age before retirement, to participate. The Gregg-Breaux plan provides personal ownership rights by allowing accumulated balances in personal accounts to be passed on to heirs, but it limits the number and type of investment options to those similar to the ones provided by the federal Thrift Savings Plan. The bipartisan Senate proposal finances its transition to individual accounts by a combination of reductions in the formula for current law OASI benefits, adjustments in Consumer Price Index calculations, and additional retirement benefit offsets.

Sen. Gregg’s June 9 testimony before the Ways & Means Committee emphasized that advance funding of future retirement benefits through individual accounts is necessary to reduce pressure for all forms of future tax increases, beginning as early as 2014 (when OASI payroll tax revenues and income taxes on retirement benefits begin to fall short of program outlays). After pointing out that large OASI Trust Fund balances have nothing to do with the federal government’s ability to pay retirement benefits, Gregg urged reductions in the amount of unfunded future benefits that would require support from general tax revenues and unfairly burden future generations. Hence, the real payoff in tax cuts from the Gregg-Breaux version of Social Security reform is not centered on future payroll tax rate levels. The plan primarily aims to reduce the general revenue hikes that will otherwise be needed to redeem OASI Trust Fund bonds, beginning in less than 15 years.

After capturing the rhetorical offensive with this positive, tax-cutting framework, the bipartisan Senate plan stumbles somewhat in its selection of policy tools. Gregg-Breaux adopts the standard measures for trimming current law OASI benefits — accelerated increases in the normal retirement age (with additional indexation for changes in future life expectancy) and unspecified actuarial adjustments for early retirement. Although the plan’s proposal to increase the period of wage earning years for which benefits are calculated is advertised as a step to reward work, its net effect on average indexed monthly earnings in the OASI benefit formula could well reduce initial payments to many future retirees.

Partly because Gregg-Breaux needs additional financing to carry out its other income redistribution goals (higher guaranteed minimum retirement benefits, increased survivors’ benefits for widowed spouses, taxpayer matches of additional voluntary contributions of lower income workers to personal accounts, annual gifts to young children through additional KidSave accounts), it over-reaches by requiring that future inflation adjustments be set 0.5 percent lower than current Consumer Price Index calculation methods provide. Besides arbitrarily lowering OASI benefit payments to future retirees, this measure also would try to collect more income tax revenue for Social Security through hidden bracket creep (higher marginal income tax rates as workers’ nominal income rose). The plan also proposes several other modest revenue grabs to assist the retirement program’s balance sheet, such as changing the indexing of the cap on taxable wages and reclaiming from Medicare the latter’s share of income taxes on retirement benefits.

Gregg-Breaux wisely does avoid the clutches of “clawback” proposals, allowing workers to benefit fully from the accumulated savings and investment earnings in their personal security accounts. At the same time, it manages to stake out new territory in recommending one possible approach to “benefit offsets.” Gregg-Breaux recognizes the need to adjust traditional OASI benefit promises for workers who will fund personal accounts with diverted payroll taxes. However, its offset formula aimed at making this adjustment equal to the “present value” of the refunded taxes misfires.

The Senate bipartisan plan uses the OASI Trust Fund interest rate as a discount factor. It confuses this rate — which is used for intragovernmental budgetary crediting of “earnings” on the paper IOU assets held in the Trust Fund — with the more appropriate rate at which taxable wage earnings (and the payroll taxes levied on them) increase OASI benefits on the margin. The latter rate must be based on the current law retirement benefits formula, which produces various “rates of return” on workers’ earnings. Those rates of return depend on a worker’s wage level, his earnings history, and the national economy’s rate of real growth in average wages during his working lifetime. Although the Gregg-Breaux benefit offset may be “close enough for government work,” it will tend to reduce future benefits relative to wage carve outs more than the current law benefits formula indicates. That feature may undercut its ability to rebut charges that the offset simply imposes politically driven benefit cuts to find budget savings (rather than economically neutral adjustments reflecting the level of payroll tax carve outs).

Gramm Plan

The Social Security Preservation Act, sponsored by Sen. Phil Gramm (R-TX), provides a higher initial level of contributions to fund a new Social Security Savings Account for Employees (SAFE Account) and a more generous minimum benefit guarantee than the above proposals. However, its suggested methods for maintaining Social Security program solvency and financing the transition to an investment-based alternative retirement plan are largely cosmetic, if not illusory. They rely heavily on general revenue transfers and provide no net additions to national saving for more than 30 years, if ever.

The Gramm plan outlined before the Ways & Means Committee on June 9 provides a carve out from Social Security payroll taxes. It allows participating workers to invest three percent of their OASI-covered wages into individual SAFE accounts. According to an earlier plan description, Gramm’s proposal would allow current workers to retain the option to join the new retirement system “at any point during their working lives.” The latter provision might slow initial participation rates by risk-averse workers, but the Social Security Preservation Act provides other strong incentives to set up SAFE accounts. Gramm guarantees all participating workers that their SAFE benefits will be no less than current law OASI benefits, plus an additional bonus equaling 20 percent of the annuitized value of their individual SAFE accounts.

However, this generous guarantee increases the transition financing burden for the Gramm plan. It reinforces a political commitment to make no changes in current law OASI benefit levels. The guarantee to participating workers also means that they will have less to lose, or gain, from their investment choices for at least several decades. While not exactly a “reverse clawback,” the guarantee provision sets another unfortunate precedent of one-way capitalism (“Heads, I win. Tails, taxpayers lose”).

The Gramm plan first acknowledges a “cash flow problem” that might last for more than 30 years in moving from current law OASI to an investment-based retirement system. Moreover, Gramm projects that the full transition to a net surplus position for the overall post-reform Social Security system would not occur until 2053. However, the Social Security Preservation Act attempts to sidestep and deflect those fiscal challenges with several accounting tricks and shaky budget assumptions.

Gramm’s testimony before the Ways & Means Committee referred obliquely to “allowing” workers age 35-55 in the year 2000 to “invest” an extra two percent of their wages in SAFE accounts. But those extra investments will be used entirely to fund current law OASI benefits, and they would not be used to calculate a worker’s 20 percent “bonus.” This not so clever device to close some of the transition finance gap is camouflaged by offsetting transfers of additional funds from projected Social Security budget surpluses over the next ten years. It either quacks like a tax hike “duck” or represents slippery double counting.

Indeed, the Gramm plan’s transition finance already relies heavily on massive revenue transfers from projected surpluses in the unified federal budget for nearly two decades. It declines to find budget savings through the various types of retirement benefits reductions proposed by several other Social Security reform plans. Instead, Gramm discovers additional magic money from “recapture” of the additional federal corporate income taxes that will be generated by SAFE account investments, as well as redemption of paper IOUs in the OASI Trust Fund.

The tax recapture device relies initially on several “heroic” assumptions and then leaves future contentious calculations to the Secretary of the Treasury. It also oversimplifies and confuses the complex linkage between the taxable income of publicly traded companies and the annual appreciation in their stock values.

The Gramm plan depends heavily on maintaining Social Security’s convoluted trust fund accounting structure as a device to cover the tracks of its financial flows and political calculations. The plan is unwilling to recommend net budget savings through reductions in unfunded Social Security promises under current law — or additional cuts in other federal expenditures (unless you believe an off-hand reference to extending current law caps on discretionary spending through 2009 represents a credible promise). Indeed, when you take the wrapper off the plan’s free lunch, one finds a very long-term commitment to asset reshuffling, general revenue financing of transition costs, and investment-based ways and means that are aimed primarily at guaranteeing existing retirement benefit promises. At another House Ways & Means Committee hearing last November, Sen. Gramm conceded that, for the “first 30 or 40 years,” almost all the benefits of new investment under his plan would have to be used to pay off the debts of the current law Social Security system.

Smith Plan

The Social Security Solvency Act, sponsored by Rep. Nick Smith (R-MI), would finance a 2.6 percent carve out for Personal Retirement Savings Accounts (PSRAs) by slowing the growth of current law retirement benefits, “by a small amount each year for a long time.” Smith points out that OASI benefits are scheduled to nearly double in real value over the next 75 years. The advantage of his “Chinese water torture” approach is that it facilitates targeting benefit cuts by workers’ age and income. Future beneficiaries from younger generations and those earning higher wage incomes, who would benefit the most from the PSRA carve out, also would absorb the greatest share of benefit growth reductions.

The Smith plan would change the current law benefits formula to slow the growth rate of benefits in the brackets for higher income workers. His proposal would add a new five-percent replacement rate for the highest paid wage earners. It then imposes annual percentage reductions in the top three replacement rates (the five-percent bracket along with the current law replacement brackets of 15 percent and 32 percent, respectively) for the portion of annual wages that would be earned only by moderate- to high-income workers. Smith also would use price indexing, rather than current law wage indexing, to adjust the bend points that separate the above replacement rates.

Future benefits would be reduced more for relatively younger generations of workers. The Smith plan accelerates the scheduled increase in the normal retirement age (NRA), raising it to age 67 by 2010. For later years, it would automatically adjust that age for unreduced retirement benefits, by indexing the NRA to life expectancy.

However, the Smith plan stretches beyond the above strategies in searching for additional sources of revenue and budget savings to restore solvency to its vision of a future Social Security program. Smith’s legislation would impose payroll taxes on newly hired state and local government workers and force them to join the Social Security retirement system. It would remove their option under current law to seek better pension alternatives offered by many of their government employers.

The Smith plan also levies a steep OASI benefits offset against the returns from PSRAs. It makes the OASI Trust Fund a silent investment partner for workers contributing to the individual accounts. In effect, the traditional retirement program would recapture the first 3.7 percent real rate of return on PSRA contributions. This offset provision severely limits the positive incentives and tangible rewards of investment-based personal accounts. Workers will need to wait even longer to realize any significant upside gain from this version of Social Security reform, which sets fiscal solvency as its highest priority. They well could feel like they were merely borrowing their own money from the government (a secured lender, as it were) — with the obligation to repay it at a high rate of interest before they could hope to capture any investment gains for themselves.

The Smith plan sends the “green eyeshade” message that achieving and sustaining a positive budget balance for the OASI Trust Fund is more important than improving the income levels and financial independence of future retirees. Although its aversion to phony accounting tricks is laudable, the Smith plan is caught in a precarious balancing act as it tries to walk a narrow tightrope between the retirement program’s annual revenues and its outlays. It takes very small and measured steps, instead of bold leaps to inspire the imagination and dreams of younger generations of workers.

Sanford Plan

The USA Accounts Act, sponsored by Rep. Mark Sanford (R-SC), offers a different approach to incremental reform of Social Security. It proposes that new USA accounts for working Americans be funded from general revenues, based on a rebate each year of their proportionate share of the entire annual surplus in the OASI Trust Fund — the excess of annual OASI revenues over the program’s expenditures. This “virtual” carve out would be financed by several types of reductions in current law OASI benefits.

Future retirement benefits for high-wage workers would be trimmed gradually. The Sanford plan would reduceonly the portion of those benefits that those workers gain by earning lifetime average covered wages that are above the second “bend point” in the current law benefits formula. (That component of wage income already is subject to a 15-percent replacment rate. Beginning with individuals first becoming eligible for benefits after 2010, this portion of their initial benefit amount will be reduced 2 percent a year under the USA Accounts Act. For individuals who become eligible in later years, the overall reduction percentage is based on the difference between their initial year of eligibility and the year 2010, as multiplied by the 2-percent annual reduction rate. The maximum benefits reduction percentage for the high-wage slice of covered earnings is set at 38 percent, and it would apply to individuals first becoming eligible for benefits in 2030 or later.)

The USA Accounts Act completes its transition finance plan by counting the rebates of OASI Trust Fund surpluses to workers as “early payment of future benefits.” The plan would provide an offset against their future OASI benefits, by subtracting the present value at retirement of the total rebated contributions that were made to a worker’s USA Account. The proposed offset formula would use the average annual yield on Treasury bonds held by the OASI Trust Fund as its discount factor for this offset calculation.

The Sanford plan strives to increase the progressivity of its retirement benefit structure with both traditional income redistribution measures and more subtle design features. It would help workers making less than $25,000 a year to build up the balance in their new USA Accounts by providing, in each of the first two years, refundable tax credits to supplement the rebates they receive. Sanford’s proposal also would exempt low-income workers from any future OASI benefits offset. The plan’s best type of targeted subsidy, however, involves its use of general revenues to pay the administrative fees of USA accounts for workers earning less than $20,000 a year. Most importantly, because the Sanford proposal explicitly chooses not to mandate the purchase of retirement annuities, it would decrease the current OASI system’s linkage between life expectancy and benefits received, thereby favoring the many lower income workers who experience shorter-than-average life spans.

The USA Accounts Act would strengthen efforts to not merely preserve projected OASI Trust Fund surpluses in a Lock Box, but actually utilize them both to enhance retirement income security and reduce OASI’s future unfunded liabilities. However, its carve out of annual OASI Trust Fund surpluses overstates their magnitude, because the bill’s calculations of this amount would include the annual interest earnings on Trust Fund “assets” that are credited to OASI under current budgetary rules. The Sanford plan also lacks a clear mechanism for drawing down equivalent amounts of future budget authority from the OASI Trust Fund itself (to fully reduce unfunded liabilities). Thus, it runs the risk of further accounting confusion and double counting of apparent surpluses in later years. Finally, the plan’s benefits offset formula, similar to the one in the Senate bipartisan plan, mistakenly uses the OASI Trust Fund interest rate as a discount factor, which would reduce future benefits relative to wage carve outs more than the current law benefits formula indicates.


Those Who Cannot Remember the Past Are Condemned to Repeat It

Despite the above flickers of congressional activity, immediate prospects for comprehensive overhaul of Social Security appear dim. Even reform-minded legislators remain tethered to static projections of federal budget balances and the distortions of trust fund accounting. Their incremental proposals focus too much on preserving budget solvency within a flawed defined benefit structure and too little on accelerating the transition to a market-based approach that will enhance economic growth and boost retirement income. In this way, well-intended reform efforts threaten to lose their sense of direction and wander one step backward.

Reluctance to move fully away from the current law baseline of promised retirement benefits is reflected in the level of minimum guarantees offered by a number of reform plans. Partial privatization of Social Security that is premised on one-way capitalism (the lure of upside gains with protection against downside risk) undercuts and confuses the core reform message: workers should rely more on the long-term incentives of market-based investment, and less on the shaky short-term promises of politically driven transfer payments.

Combining personal control and ownership of defined contribution investments and their contingent returns with a corresponding set of fixed defined benefits guaranteed by taxpayers represents an instable and contradictory mix. It quickly gives rise to political controls on the scope and scale of private investment choices (e.g., narrow portfolio options, limits on changes of investment advisors, mandatory annuitization at retirement), both to protect against market risk and to limit moral-hazard-driven losses. The tension between competing desires – to maintain guaranteed benefits and to allow workers more control over investment decisions – is not easily resolved.

On the other hand, the potential of significant investment gains from funds that receive taxpayers’ contingent guarantees will trigger parallel efforts to capture a share of them for politically determined income redistribution objectives. In short, politicians will argue that placing taxpayers’ funds on the line buys them a partnership share in monitoring and guiding personal retirement account investments. But if workers lose control over investment and consumption of their “privatized” assets, they will forfeit the potential benefits of vigorous market competition and question the value of mandated savings. Social Security modernization efforts will stumble another step backward.

Corrective Lenses, For Better Vision

At this point, the first step forward begins with reframing the budgetary treatment of Social Security’s retirement commitments in order to clarify voters’ full range of policy choices and rise above today’s quagmire. For example, potential opportunities and necessary tradeoffs cannot be debated adequately without improved understanding of the illusions of trust fund accounting. As noted recently by Federal Reserve Bank of Cleveland economists David Altig and Jagadesch Gokhale, the OASI Trust Fund extends the tax base for financing Social Security benefits to include future income taxes. The device of accumulating paper IOUs in the trust fund “simply permits current generations to stake a claim on future generations’ general tax payments while reaping the benefit of current government spending on public goods and services financed by Social Security surpluses,” they conclude. Beneath the budgetary smoke and mirrors of trust fund accounting, the IOUs in the OASI account represent promises to pay a rising share of prospective retirement benefits with future general revenues collected from younger generations.

Financial journalist Martin Mayer once quipped about the balance sheets of troubled banks that “the assets [shaky loans] are liabilities and the liabilities [bank deposits] are assets.” Converting the steady “deposits” of workers’ payroll taxes into more valuable assets, of course, still depends on allowing them to be invested productively through personally owned and controlled retirement accounts. In the meantime, trust fund accounting understates the long-term unfunded liabilities of the OASI program and overstates its assets in several respects.

Current accounting practices for the federal budget treat the IOUs held by OASI as it they were real assets, instead of future claims against taxpayers. Then they credit the OASI Trust Fund with notional interest “earnings” on those assets, through intragovernmental transfers of future budget authority based on a blended mix of longer term Treasury bond yields. Finally, the trust fund accounting used by the OASI Board of Trustees limits its estimates of future program solvency, as a percentage of taxable payroll, to a 75-year window even as it presumes a perpetual set of benefit commitments. However, the present value of Social Security’s Trust Fund deficits on a perpetual basis is at least twice as high as the 75-year estimate.

A more accurate portrayal of the fiscal pressures on Social Security would also highlight the inherent limits on future federal government claims on the goods and services produced by American workers. Rebalancing the unfunded obligations of the current retirement system will impose substantial costs to taxpayers and/or beneficiaries, relative to current law. Utilizing general revenues to finance those transition costs, even in the midst of unified budget surpluses, does not eliminate this burden. It may help avoid deeper cuts in promised retirement benefits, but the opportunity costs of that policy decision merely shift to other sectors of federal budget policy. Competing government functions increasingly become strapped for dollars, and other federal policy alternatives such as tax cuts or new program initiatives are foreclosed.

The best way out of this policy box is to free additional resources for greater private saving and more aggressive Social Security reform by restraining the growth of overall federal spending. The most sustainable case for long-term restructuring of retirement benefit financing must first emphasize the superior gains offered by advance funding of individually owned and controlled investment accounts. But it then needs to ensure that this funding is built on greater net amounts of saving, rather than a reshuffling of public debt obligations. Thus, the issue must come down to workers choosing between the value of marginal government services and the greater retirement income security that federal spending reductions would make possible.

Unfortunately, this year’s political climate finds no major congressional figure stepping forward to advocate substantial cuts in corporate welfare, elimination of “dinosaur” agencies and programs, and a pork-free diet of transportation spending in order to accelerate advance funding of an individual account-based retirement system. A full-fledged offensive to overhaul Social Security comprehensively apparently will require greater public education and the advocacy of a successful presidential candidate.

One Step in the Right Direction

However, in the event that near-term incremental efforts at Social Security reform inch forward on Capitol Hill, they should avoid doing harm to ultimate objectives. Today’s political timidity in addressing the embedded unfunded liabilities within the current set of retirement benefit promises may place a low ceiling on the size of individual account carve outs. It need not preclude structuring those accounts in a manner consistent with a later move to full-scale overhaul of the retirement program.

Thus, any single step forward must transcend the narrow limits of today’s federal-budget-centered debate and its trust fund accounting illusions. It also must move beyond backward-looking efforts to re-guarantee the inadequate benefits promised under current law. To avoid squandering the current window of operating room provided by budget surplus projections, a successful incremental measure should use the revenue cushion provided by current OASI payroll taxes to advance fund an opening round of individual investment accounts. But this first stage of Social Security reform should reinforce the long-term message of market-based investment and personal saving — workers will have greater opportunities to seek higher returns on savings and more retirement income by accepting the market risks of more volatile private investments, in exchange for reducing their claims on younger workers and future generations of taxpayers through fixed benefit guarantees.

Hence, sustaining this initial step toward a market-based personal saving approach to retirement income security will hinge on applying a credible method for reducing a portion of the defined benefits guaranteed under current law. The appropriate benefits offset formula will adjust future OASI outlays in proportion to the level of payroll tax revenues that are carved out of the current OASI program, and thereby begin the lengthy process of first capping and then rolling back the retirement system’s enormous sum of unfunded liabilities.

The proper benefits offset would involve removing a portion of a worker’s taxable wages from the OASI earnings history that is used to calculate his initial monthly benefit, in the same proportion as the size of an individual account carve out relative to total OASI payroll taxes under current law. To illustrate, assume that the individual account carve out is 2.65 percent of taxable payroll (one-fourth of the current law OASI payroll tax rate, which will be 10.6 percent in 2000). Thus, if a worker earned $20,000 in 2000, he only would be credited for OASI wage history purposes with earning ¾ of that amount, or $15,000. The other $5,000 of his “taxable” earnings would be diverted out of OASI to support his contribution of $530 into an individual account. (Although the remaining $1,590 in payroll tax revenue would have been used to pay current OASI benefits for other retired workers, the $15,000 in wages from which it was collected still would be credited toward calculation of that worker’s future OASI benefits. Despite the obvious strains of this intergenerational double counting, that’s the Social Security Administration’s basic Ponzi story, and they’re sticking to it).

Ideally, a new retirement system based partly on a personal account carve out and partly on traditional OASI benefits would make the above adjustment at the front end of retirement benefit calculations (i.e., the amount of wages that are posted to the OASI earnings history account) to reflect the appropriate percentage of a carve out. Then all of the necessary backend benefit reductions would take place without having to impose any political judgments on how much to cut, in what “sneaky” manner, and involving which victims who won’t be able to squeal as loudly as someone else. No one would get a penny less than they truly were “entitled” to receive under the current law benefit formula, once you allowed for their choice to redirect a portion of their taxable wages into an individual account carve out.


It remains unlikely that sufficient members of the current Congress will muster the backbone to deal with the long-term unfunded liabilities of the current law retirement benefits structure. But in the meantime, the approach proposed above for starting a personal account carve out from OASI revenue surpluses, in combination with OASI benefits offsets, would offer a straightforward option to workers. If they want to try to earn more money on their retirement “savings” through individual account investments, all they have to do is reduce the appropriate amount of their earnings that otherwise would be used to “build” future Social Security defined benefits, and then direct them instead to the contingent returns of individual account investments. No one would get anything less than the defined benefit returns on whatever they chose to put in to the “old” OASI benefit formula, but they would be guaranteed nothing more.

This year’s stunted debate over Social Security restructuring has lowered expectations. It suggests that the most appropriate short-term goal now is to avoid enactment of flawed or misleading proposals. Of course, the ebb and flow of upcoming presidential campaigns again could rearrange the policy cards and open up broader opportunities for reform. In the meantime, however, there is little excuse for pre-compromising away the fundamental tenets of sustainable retirement system reform and moving the terms of debate two steps backward, instead of one step forward. Even without any clear signs of concrete action, that kind of talk is not cheap. It could prove to be quite expensive.